README.md

August 28, 2026 · View on GitHub

dsh-claude-compat

npm downloads/month npm total downloads

Reuse your entire Claude Code setup in DSH — zero migration. skills, commands, rules, agents, hooks, MCP — install and they just work. Nothing to reconfigure.

The 10-second pitch

You've already built up a solid .claude/ in Claude Code — skills, slash commands, rules, agents, hooks, MCP servers. Moving to DSH (DeepSeek Harness) shouldn't mean rebuilding all of it.

dsh-claude-compat is that bridge. It carries your .claude/ directories verbatim into DSH — everything works immediately. Install once, and both sides share one config: change it in one place, it updates in both.

dsh plugin --profile web add dsh-claude-compat@latest
dsh web

Done. Your skills show up in the / menu, rules inject into every new session, and you can invoke agents with /agent-name — all inside DSH, exactly like Claude Code.

Or let DSH install it itself (recommended): copy this line to DSH — it reads the skill, installs, and verifies:

Read https://github.com/biedongbin/dsh-claude-compat/blob/main/.claude/skills/install-dsh-compat/SKILL.md and install the dsh-claude-compat plugin following its steps.

⭐ Find this useful? Give it a star — it keeps us improving and helps more people discover it.

What it brings over — at a glance

Your Claude Code asset	Becomes in DSH	How to use it
`skills/**/SKILL.md`	DSH skill (lazy-loaded)	`/skill-name`, the `skill` tool, and the model-visible catalog
`commands/*.md`	DSH skill (user-invocable)	`/command-name` in the slash menu
`rules/*.md`	message-stream injection (same channel Claude uses)	auto-applies to every new session
`agents/*.md`	delegation-shim skill	invoke `/agent-name` with its persona
`.claude/settings.json` hooks	Pre/PostToolUse + UserPromptSubmit bridged	commands, permissions, blocks behave as before
`<root>/.mcp.json`	`dsh-mcp-client` instances	stdio / streamable-http auto-translated
`~/.claude/plugins`	DSH skill (managed via `/cc-plugin`)	installed plugin skills work too

Both the project .claude/ and the user-level ~/.claude/ are read. Same-name entries dedupe by fixed priority: project .claude > DSH native > ~/.claude — a project skill always wins over other copies, and a user skill never overrides a DSH-native one. CLAUDE.md / AGENTS.md are handled by DSH's built-in dsh-agent-instructions — this plugin never touches them.

What it does (implementation detail)

`.claude/` path	Mechanism	Behavior
`skills/**/SKILL.md`	DSH skill provider	Name + description in the model-visible catalog; body loads on demand via the `skill` tool. New skills appear on the next catalog reconcile — no restart.
`commands/*.md`	DSH skill provider	Same, plus user-invocable: `/command-name` works in the slash menu.
`rules/*.md`	Message-stream injection	Rules are concatenated, wrapped in a `<system-reminder>` envelope, and prepended as a user-role message at the front of the message array once per session — the same channel Claude Code uses (`prependUserContext`), which models follow reliably.
`agents/*.md`	DSH skill provider (delegation shim)	DSH has no markdown subagent format, so each agent file becomes a skill whose body leads with an explicit "delegate with this persona" instruction. Model- and user-invocable, so `/agent-name` works. Same-name agents dedupe by rank like skills.
`.claude/settings.json` → `hooks`	Tool/prompt hooks	Claude Code hooks subset bridged onto DSH's `tools/pre-execute` (PreToolUse), `tools/post-execute` (PostToolUse) and `agent/pre-step` (UserPromptSubmit) waterfalls. Commands run via `/bin/sh -c` with a Claude-style JSON payload on stdin; exit code 2 denies/blocks, `hookSpecificOutput` overrides are honored, timeouts allow through with a warning.
`<projectRoot>/.mcp.json`	MCP servers	Claude Code-format MCP server definitions are translated to `dsh-mcp-client` plugin instances at DSH startup from the launch workspace: `command` → stdio, `url` → streamable-http. Malformed entries or a missing `@deepseek-ai/dsh-mcp-client` degrade to a warning, never a crash.
`~/.claude/plugins` (installed plugins)	DSH skill provider	Installed Claude Code plugin-marketplace plugins contribute their skills/commands/agents (via each install's `.claude-plugin/plugin.json` manifest, or a directory scan when manifest-less) at rank `750` — the long tail: project `.claude`, DSH native, and `~/.claude` all win collisions. Plugin MCP servers (`mcpServers` in the manifest) mount only when `enablePluginMcp` is opted in — mounting third-party MCP servers is a bigger trust step than listing skills.

The same directories are also read from the user-level ~/.claude/ (skills, commands, rules, agents, and ~/.claude/settings.json hooks). Same-name skills/commands/rules/agents are deduped with a fixed priority:

project .claude > DSH native (.dsh) > ~/.claude

Project .claude entries carry rank 50; DSH's own skills — project .dsh roots, .agents roots, and bundled skills (ranks 100–600, BUNDLED_SKILL_RANK) — sit in between; ~/.claude entries rank 700. So a project skill always overrides the DSH-native and user copies, and a user skill never overrides a DSH-native one.
Rule files with the same basename in ~/.claude/rules are skipped when the project already provides one.

CLAUDE.md / AGENTS.md are not touched — DSH's built-in dsh-agent-instructions already handles those.

Built-in commands

Installing this plugin adds three management skills to the catalog:

Command	What it does
`/cc-plugin`	Full Claude Code plugin management: `list`, `install <name>[@marketplace]`, `uninstall`, `enable`, `disable`, `update [name]`, `search <term>`, `marketplace list\|add\|remove\|update`. One-shot syntax `/cc-plugin <name>@<marketplace>` installs directly. Engine: the `claude` CLI when available, otherwise a built-in fallback (direct JSON + git, with timestamped backups of every file it touches). All state stays in Claude-native locations (`~/.claude/plugins`, `~/.claude/settings.json` `enabledPlugins`) so Claude Code and DSH read the same truth.
`/reload-cc-plugins`	Hot-reload the skill catalog: drop cached provider lists and notify observers so newly installed/removed skills appear in the current session — no restart, no new session.
`/reload-skills`	Alias of `/reload-cc-plugins`.
`/cc-export`	Export DSH-native skills (`.dsh/skills`) into Claude Code `.claude/skills/<name>/SKILL.md` with frontmatter preserved. `list` / `export [--overwrite] [--target]`.
`/cc-resume`	List Claude Code conversation sessions for the current project (`~/.claude/projects/`) and import any of them into DSH with full user/assistant/tool history. Imported sessions appear in the DSH session list titled `cc: <preview>` and resume like native ones. `list` / `import <sessionId>` / `--limit-turns N` for huge sessions.

Typical loop: /cc-plugin install ralph-loop@claude-plugins-official → /reload-cc-plugins → new skills visible immediately. Plugin-shipped MCP servers still require a DSH restart (process-lifetime mount).

Requirements

DSH with a profile (e.g. web)
A project using Claude Code conventions: .claude/skills/, .claude/commands/, .claude/rules/, .claude/agents/, .claude/settings.json, and a project-root .mcp.json (all optional; ~/.claude/ equivalents are also picked up)
pnpm on PATH — dsh plugin is a thin pnpm forwarder

Install / Update

One command — the package declares dsh.bundle, so DSH activates it automatically (no manual cordis.patch.yml editing). Install or update to the latest release:

dsh plugin --profile web add dsh-claude-compat@latest

Or from GitHub:

dsh plugin --profile web add github:biedongbin/dsh-claude-compat

Restart DSH (dsh web). Done — skills show up in /, rules are injected into every new session.

Configuration

Option	Default	Description
`enableSkills`	`true`	Register the `.claude/skills` + `.claude/commands` + `.claude/agents` provider (project and `~/.claude`)
`enableRules`	`true`	Inject project + `~/.claude` `rules/*.md` into the message stream
`enableMcp`	`true`	Translate `<projectRoot>/.mcp.json` into mounted MCP server plugins
`mcpFailOnStartupError`	`false`	Forward to `dsh-mcp-client`: fail plugin startup when an MCP server fails to connect
`enableHooks`	`true`	Run `.claude/settings.json` hooks (Pre/PostToolUse, UserPromptSubmit)
`hooksTimeoutMs`	`60000`	Per-hook run timeout (UserPromptSubmit capped at 10s regardless)
`enableAgents`	`true`	Surface `.claude/agents/*.md` as delegation-shim skills
`rulesMaxBytes`	`65536`	Hard cap on total injected project rules text
`userRulesMaxBytes`	`65536`	Hard cap on total injected `~/.claude/rules` text
`projectRootMarkers`	`[".git"]`	Ancestor markers for project-root discovery
`skillRank`	`50`	Provider rank for project `.claude` skills (wins every DSH-native collision)
`skillSource`	`project-claude`	Source tag for project catalog entries
`userSkillRank`	`700`	Provider rank for `~/.claude` skills (loses to DSH-native `600`)
`userSkillSource`	`user-claude`	Source tag for `~/.claude` catalog entries
`userClaudeDir`	`~/.claude`	User-level `.claude` directory (`~` expands to the home dir)
`enablePlugins`	`true`	Surface skills/commands/agents from installed Claude Code plugins (`~/.claude/plugins`)
`pluginSkillRank`	`750`	Provider rank for plugin content (the long tail — everything else wins)
`pluginSkillSource`	`claude-plugin`	Source tag for plugin catalog entries
`pluginsRoot`	`~/.claude/plugins`	Plugin-marketplace root (`installed_plugins.json` + `cache/`)
`enablePluginMcp`	`false`	Mount plugin-declared MCP servers (opt-in; requires `enablePlugins` and `enableMcp`)
`enablePluginManager`	`true`	Register the `/cc-plugin`, `/reload-cc-plugins`, `/reload-skills` management skills
`pluginManagerRank`	`40`	Rank for the built-in management skills (top of the catalog)

Notes

Skill naming: DSH requires kebab-case skill names. Nested skill directories are flattened (gitnexus/gitnexus-guide → gitnexus-gitnexus-guide); invalid frontmatter names fall back to the directory name.
MCP lifecycle: .mcp.json is read once at DSH startup from the launch workspace — not per session — and each server mounts for the process lifetime. Restart DSH to pick up edits.
Hooks scope: a deliberately small subset of Claude Code hooks: PreToolUse / PostToolUse / UserPromptSubmit. Matchers support exact names, * wildcards, and | alternation; commands run with stdin carrying the Claude-style JSON payload. Exit code 2 = deny (Pre) / block (Post); other non-zero exits and timeouts allow through with a warning.
Rules granularity: rules are read per new session (cached per session cwd). Editing a rule mid-session takes effect in the next session.
Rules content: rules are injected verbatim as instructions to the model. Only commit rules you want the model to follow — same trust level as CLAUDE.md.
Catalog snapshot timing: the skill catalog is snapshotted when a session is created. Skills installed or edited mid-session surface after /reload-cc-plugins (hot reload) or in the next session.

Troubleshooting

Known limitation: `/cc-resume` via the skill tool

In some DSH runtime configurations the skill tool resolves in an agent-scoped layer where globally registered providers are not visible — the invocation returns skill "cc-resume" is unknown or no longer available even though the catalog lists it. This is a DSH runtime layering behavior, not a plugin bug. The skill body only instructs the model to run the CLI, which is always available:

node node_modules/dsh-claude-compat/scripts/cc-resume.mjs list
node node_modules/dsh-claude-compat/scripts/cc-resume.mjs import <sessionId>

DSH won't start back up after a restart / port 3080 stuck. Old process still holding the port (symptom: EADDRINUSE in logs). Use the bundled restart script — it waits for a clean stop, falls back to SIGKILL, and verifies the port before reporting success:

npx dsh-claude-compat-restart        # bin alias (installed with the package)
bash node_modules/dsh-claude-compat/scripts/dsh-restart.sh   # direct
bash scripts/dsh-restart.sh --no-patch   # skip the prompt patch, restart only

The script also re-applies the idempotent dsh-terminal-bash prompt patch, which npx/npm updates silently revert. DSH_RESTART_PORT overrides the port (default 3080).

Installed a plugin via /cc-plugin but its skills don't show. Run /reload-cc-plugins. Still missing → restart DSH (plugin-shipped MCP servers always need a restart).

/cc-resume import fails on compression. The importer needs the zstd binary (macOS: brew install zstd; most Linux images ship it).

/cc-plugin reports "claude CLI unavailable". The fallback engine handles install/enable/disable; for marketplace add/update, install Claude Code (npm install -g @anthropic-ai/claude-code) or manage marketplaces from Claude Code directly.